Serious gainclone problem

[firefly]

Hello all

some time ago I built Joe Rasmussens Tube Gainclone ( http://www.customanalogue.com/diytub...lone/index.htm )
The amp worked fine for some weeks, until I had the unfortunate idea of pushing the amp a little too hard (well I was listening to Keith Jarrett and you know.....)
It made a THUMP and THUMPTHUMPTHUMP and I thought OK I fried the IC's and have DC at the output.
I changed the IC's (I wrongly assumed they were gone), improved the heatsinking and..... nothing changed, the amp THUMPed.
I made some measurements with dummy loads (they get hot!!) and I have DC at the output that slowly (4-5 sec.) "oscillates" from +12VDC to -12VDC.
Fried Transformer? Coupling capacitors? Any ideas where I should start looking?
Thanks!

[Nuuk]

Well, first "serious" is what's happened in Burma just to put things into perspective.

What are your offset readings without the valve buffer attached?

[firefly]

Hi Nuuk

yes... I might have been overreacting

I pulled the tube out and..... no DC! Well I get a spike when I turn the amp on but then it quickly settles to 15mV, I assume that's like it should be.
I put the tube back in and measured if there is DC after the coupling capacitor but there's none.
I'm perplex there is a lot I don't understand!
Is there a next step? Thanks!

[Nuuk]

Yes, that is odd if you have no DC after the coupling cap but the problem is solved by taking out the valve.

Do you have another valve you can try?

[firefly]

I'll have to order one and test again.
I tried another transformer but the problem remains (at least I know it's not the transformer...)
the fastest solution would be to change all the components... but it could happen again and that's not the goal....
it still puzzles me what component could have been "forced"... the tube seems odd (I used a new EH 6922), but I'll try as soon as I get a new one.

what do you think about using a pcc88? there's plenty around, NOS european and american brands and you can get them for less then an ecc88 or e88cc.... and the heater is easily adapted with a different resistor.

[djk]

Dos it thump about 4~5 times per second?

[firefly]

Hi djk

no, I'd say it thumpes 4 times in 4 secs, I guess it works like this:
DC goes up quickly to 12V (thump1) then decreases quickly (thump2). Decreasing slows down when getting near 0V. Then up again quickly to -12V (thump3) and so on...

I should get a new tube this week.... I'll see

[firefly]

Haven't got the tube yet....
but I made some more measurements:
when I switch on without the tube I get this DC spike before it settles on 15mV. What I didn't notice before is that it behaves exactly like when the tube is in, DC goes up to 12V (and the dummy load gets hot) and decreases quickly. The only difference is that without the tube DC settles on 15mV instead of going up to -12VDC.

So I cut the PSU from the tube circuit and: No spike at switch on and DC ~3mV.
At this point I don't think it's the tube.

I have the feeling it must be something obvious but my knowledge is too limited

Any ideas?

[Nuuk]

You will get a high DC offset while the tube buffer 'settles down'. That's why I used a speaker protection module that delayed connecting the speakers to the amp for a second or two. Another option is to use some sort of output delay on the output of the tube buffer.

Of course, that is probably nothing to do with your problem. Did you copy the Joe Rassmussen circuit exactly?

[AndrewT]

fit a delayed input unmute and output disconnect relay to give time for the amp to settle.
Is the amp stable into an open circuit load? You may have to relay such that a suitable load is switched in during start up.